Skip to content

Add optional O parameter to PyMatching plugin#449

Merged
bmhowe23 merged 5 commits intoNVIDIA:mainfrom
bmhowe23:pr-decode-to-obs
Apr 10, 2026
Merged

Add optional O parameter to PyMatching plugin#449
bmhowe23 merged 5 commits intoNVIDIA:mainfrom
bmhowe23:pr-decode-to-obs

Conversation

@bmhowe23
Copy link
Copy Markdown
Collaborator

Add optional O parameter to PyMatching plugin to decode to observables instead of decoding to errors.

Signed-off-by: Ben Howe <bhowe@nvidia.com>
Signed-off-by: Ben Howe <bhowe@nvidia.com>
- If < 64 observables, then use decode_detection_events_for_up_to_64_observables
- Move where `result` is initialized to avoid unnecessary work when decoding to observables.
- Introduced a debug option to measure decode times using the PERFORM_TIMING macro.
- Added timing measurements at various stages of the decoding process.
- Updated the destructor to print decode times for analysis.

This enhancement allows for better performance analysis of the decoding process in the PyMatching library.

Signed-off-by: Ben Howe <bhowe@nvidia.com>
…bles

Signed-off-by: Ben Howe <bhowe@nvidia.com>
@bmhowe23
Copy link
Copy Markdown
Collaborator Author

bmhowe23 commented Apr 10, 2026

Running extra tests since this is a late change:

@bmhowe23 bmhowe23 merged commit 84d18ca into NVIDIA:main Apr 10, 2026
60 of 79 checks passed
@bmhowe23 bmhowe23 deleted the pr-decode-to-obs branch April 10, 2026 17:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant